Endoscopic nasobiliary drainage-based saline-injection ultrasound: an imaging technique for remnant stone detection after retrograde cholangiopancreatography

Abstract

Abstract Background The purpose of this retrospective study aimed to assess the accuracy of detection of remnant common bile duct (CBD) stones by injecting saline through endoscopic nasobiliary drainage (ENBD) tubes under transabdominal ultrasound (US) guidance. Method Stone extraction and ENBD are regularly achieved through endoscopic retrograde cholangiopancreatography (ERCP) in patients with CBD stones. At 1–3  days thereafter, routine US studies were performed and repeated, using ENBD tubal saline injections (20–100  mL). Results A total of 302 patients underwent standard ERCP stone extractions in conjunction with occlusion cholangiograms, routine US testing, and ENBD-based saline-injection US exams. By occlusion cholangiogram, remnant stones were suspected in 31 (10.3%) patients in total of 302, and 26 (83.8%) were verified as true positives (sensitivity, 50.9%; specificity, 98.0%). Routine US studies proved suspicious in 13 (4.3%) patients in total of 302, and 12 (92.3%) were verified as true positives (sensitivity, 23.5%; specificity, 99.6%). Using ENBD-based saline-injection US, suspected stones were identified in 50 (16.6%) patients in total of 302, and 46 (92%) were verified as true positives (sensitivity, 90.1%; specificity, 98.4%). The sensitivity of ENBD-based saline-injection US significantly surpassed that of occlusion cholangiogram (p < 0.001) and routine US (p < 0.001). Conclusion Detection of remnant CBD stones via ENBD-based saline-injection US is a valid, inexpensive, and repeatable means of patient screening that is non-invasive, radiation-free, and dynamically informative. This may help improve the accuracy of detecting remnant CBD stones after ERCP.